SAM C20/C21 availability (need 5V tolerant I/O)

Go To Last Post
17 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Anyone know when SAM C21s will become available? The Xplained boards are around and they were announced in October 2015, but the only place I can't find anywhere that stocks the parts. I'm based in the UK if it matters. Mouser lists them as non-stock items only available in reels.

 

Any ideas on pricing and availability? Or alternatives for parts with 5V tolerant I/O? I'm very restricted on board space and BOM cost, but also need a 32MHz clock and ideally DMA. I have working XMEGA code but they are 3.6V max, AVR32 is expensive and the only 5V tolerant ARM parts are not available.

This topic has a solution.
Last Edited: Tue. May 17, 2016 - 12:02 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm starting to worry that Atmel is cancelling parts now that they are owned by Microchip...

This reply has been marked as the solution. 
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

they were announced in October 2015

It would be "typical Atmel" to announce things about a year before chips are available at the usual distributors.  ATmega328pb was announced in November 2014, and people started to report it showing up at Digikey in Nov 2015.  https://www.avrfreaks.net/forum/a...

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
Or alternatives for parts with 5V tolerant I/O?
dsPIC33EV

https://www.microchip.com/design-centers/16-bit

via the product selector on DMA channels (non-zero), operation voltage max (5.5V).

Am curious to see what 8-bit MCU, other than most XMEGA, have DMA.

mojo-chan wrote:
I'm very restricted on board space and ...
Would a level translation IC fit?

These are in multiple bit widths, some in very small packages, and usually have the bus hold pullup (no added pull-up resistors).

Same idea for FET gate drive IC.

mojo-chan wrote:
AVR32 is expensive ...
That's relative; about 10USD each for 100 for the QFN.

http://www.atmel.com/devices/AT32UC3C2256C.aspx

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Cypress ARM-core PSoC chips are 5V capable.  The "SoC" stuff is a bit complicated and scary...

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Silego GreenPAK might be less onerous.

Its voltage reference is almost 1% (better than an XMEGA E) and one has a DAC that's tenth the speed of an XMEGA E DAC.

0.4mm pitch QFN.

Software on Linux, OS X, and Windows.

Silego Technology

GreenPAK: Programmable Mixed-signal Matrix

Dual Supply GreenPAK

http://www.silego.com/products/greenpak_dualsupply.html

from

Replacing a line-level converter and a comparator with a 2mm x 3mm GPAK device

by

May 16, 2016

http://www.embedded.com/electronics-blogs/say-what-/4442035/Replacing-a-line-level-converter-and-a-comparator-with-a-2mm-x-3mm-GPAK-device

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gchapman wrote:

mojo-chan wrote:
Or alternatives for parts with 5V tolerant I/O?
dsPIC33EV

https://www.microchip.com/design-centers/16-bit

via the product selector on DMA channels (non-zero), operation voltage max (5.5V).

Am curious to see what 8-bit MCU, other than most XMEGA, have DMA.

 

Thanks, I went over to take a look but unfortunately I don't think they are suitable. I need operation between 3V and 5V, and the 5V dsPIC parts only seem to go down to 4.5V.

 

Sorry, I should state my requirements more clearly since this seems to have turned into an alternative suggestion thread.

 

I have to interface with a 2MHz custom bus, so that's 0.5uS per bit. At 32MHz that only gives you 16 instructions, but of course it's actually worse than that because the data hold period is only about 0.25uS or 8 instructions. So polling becomes difficult. It's possible to do it with cycle-exact sampling instead, but it gets hairy with variable interrupt latency. However, it's quite easy to do with DMA from port to memory on an XMEGA, and doesn't force you to have maximum priority atomic interrupts that break UART comms etc.

 

So, ideally I would like:

 

DMA

High enough clock speed for the 2MHz

2M baud capable UART, 3.3V

12x 5V tolerant I/O pins

I/O must support input and open drain output

SPI (also 5V tolerant)

Operation down to 3V

Hand solderable (TQFP etc)

 

USB and space for a USB bootloader would be nice, but not essential. My current prototype uses an XMEGA E5, but is not 5V tolerant.

 

Quote:
Would a level translation IC fit?

 

I am considering them. I also looked at I/O expanders, but one of the devices I need to interface with has a requirement of 2uS response time to stimulus, so serial interfaced expanders will be too slow.

Last Edited: Tue. May 17, 2016 - 09:28 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

westfw wrote:

Cypress ARM-core PSoC chips are 5V capable.  The "SoC" stuff is a bit complicated and scary...

 

They are, and I was hoping to stick with Atmel (already got all the tools/knowledge) and to avoid using much library code. I prefer to just hit the registers directly and the SAM C20/C21 look pretty straight forward.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:
http://www.nuvoton.com/hq/produc... These parts have been available for a few years.

 

That's an interesting part. Really nice I/O and a good datasheet. I'm just worried that a lack of DMA and maximum 24MHz clock speed might make handling this 2MHz bus a bit tricky. It's only got a little over 5 cycles from the clock line falling to the data line becoming invalid.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Id be pretty worried about high-speed bus code being ported from a 32MHz XMega to ANY ARM chip.  The ARMs (and especially the M0s) aren't nearly a good at twiddling IO bits...

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

westfw wrote:

Id be pretty worried about high-speed bus code being ported from a 32MHz XMega to ANY ARM chip.  The ARMs (and especially the M0s) aren't nearly a good at twiddling IO bits...

 

True. Do you know if that applies to DMA as well? I need to read the datasheet when I get time, or I was hoping to just try it with an Xplained board.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gchapman wrote:
Am curious to see what 8-bit MCU, other than most XMEGA, have DMA.
From a partial search only at Silicon Labs :

  • C8051F06x - 8051 single cycle, 25MHz, DMA for 16-bit 1Msps ADC, external memory to 64kB data, 5V on one 8b port.
  • EFM8UB2 - 8051 mostly single cycle, 48MHz or 50MHz, no DMA, external memory interface, 5V on almost all pins.
  • EFM8LB1 - 8051 mostly single cycle, 72MHz, DMA for 1Msps ADC, no external memory, 5V is TBD but stated as tolerant (all but 4 I/O).
  • EFM8BB3 - 8051 mostly single cycle, 50MHz, DMA for 200Ksps ADC (12b), no external memory, 5V is TBD but stated as tolerant (all but 4I/O).

https://www.silabs.com/products/mcu/8-bit/Pages/8-bit-microcontrollers.aspx

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There appear to be some 5V tolerant ARMs from Atmel, such as the AT91SAM7SE and AT91M40800. It's hard to find them in the parametric table because it doesn't let you select 5.5V maximum voltage on a GPIO, but the datasheet says 5.5V is the limit.

 

Is there any way to find a complete list of 5V tolerant parts? It's really hard to find suitable parts that are actually available.

 

I have another project that only needs to interface with 5V RAM, ideally via an external bus interface (EBI). Other requirements are access to an SD card and ideally USB.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I haven't looked carefully - are the SAMC chips essentially identical to the SAMD chips of similar numbering, EXCEPT for the 5V operation thing?

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks gchapman, I'm glad they are finally starting to ship. Time to get a dev board and start playing around.

 

westfw, that does appear to be the case.